42 results about "Discrete optimization" patented technology

The invention relates to an unmanned aerial vehicle control system and method for formation flight in a wind environment. The method comprises the following steps: generating a wind disturbance environment with a specified map scale and containing random obstacles; acquiring external image data and inertial measurement data of the unmanned aerial vehicle, and generating a grid map through point cloud extraction and conversion operation; the method comprises the following steps: modeling a spatial topological structure of an unmanned aerial vehicle cluster by adopting an undirected graph mode, and solving and generating a collision-free initial trajectory by adopting an unconstrained discrete optimization mode; optimizing the initial trajectory by adopting a task redistribution strategy, and re-planning a global target and a local target to obtain a re-planned optimized trajectory; and based on the optimized trajectory, solving by adopting a model prediction control mode to obtain an optimal control quantity for unmanned aerial vehicle bottom layer control so as to correspondingly control the motion state of each unmanned aerial vehicle and realize formation flight. Compared with the prior art, the unmanned aerial vehicle cluster can efficiently and accurately realize formation flight in an obstacle environment with wind influence.

The invention belongs to the technical field of additive manufacturing, and particularly relates to a multi-scale parallel optimization method for a lightweight structure based on lattice discrete optimization. According to the method, through multi-scale parallel optimization design, a macro structure and a micro lattice structure are described through geometric parameters of a movable deformation rod piece, an improved two-value coding parameterization method, called a BCP method for short, is adopted to solve discrete optimization of the micro lattice structure, and geometric parameters of an optimization result are easy to extract.

The invention provides a plane layout optimization method and device based on CAD drawing automatic identification, relates to the field of plane layout optimization, and solves the technical problems that only non-overlapping constraints among equipment are considered, actual space obstacles cannot be automatically avoided, and a generated layout scheme cannot be implemented. The method comprises the following steps: extracting barrier data in a layer; and through a parameter configuration mechanism, generating an equipment parameter information structural body containing the mapping relationship between the equipment and the process. And establishing a continuous optimization model containing an IPOPT solver, and obtaining an initial layout solution containing the relative orientation of each device. And based on the relative orientation of each device in the initial layout solution, introducing an obstacle rejection constraint and a device non-overlapping constraint to construct a discrete optimization model, obtaining the center position coordinates of each device through a Gurobi solver, and integrating a dynamic adjustment mechanism and an interlayer relationship optimization mechanism. And generating a new layout map for visual output. The method and the device are used in a plane layout optimization process.

The invention relates to the technical field of aircraft path planning, and particularly provides an aircraft path planning method for a space non-cooperative target. The method comprises the following steps: constructing an aircraft dynamics model and constraint conditions, and obtaining a non-convex optimal control problem; performing discretization processing on a non-convex optimal control problem through a non-equal-interval pseudo-spectrum discrete optimization model, and converting the non-convex optimal control problem into a fixed-time nonlinear programming problem about discrete nodes; according to the method, a fixed time nonlinear programming problem is convexed to form a convex sub problem, a local optimal solution of a non-convex optimal control problem is obtained through sequence iteration, under the multi-state constraint, path planning of an aircraft approaching a space non-cooperative target is achieved, the problem of no feasible solution caused by non-accurate linearization is solved, and the method is suitable for being applied to a non-cooperative target. And the convergence rate of the algorithm is accelerated.

The present application belongs to the technical field of additive manufacturing, and particularly relates to a lightweight structure multi-scale parallel optimization method based on lattice discrete optimization. The present application uses the geometric parameters of movable and deformable rods to describe macroscopic structures and microscopic lattice structures through multi-scale parallel optimization design, and uses an improved double-value coding parameterization method, referred to as BCP method, to solve the discrete optimization of the microscopic lattice structure and easily extract the geometric parameters of the optimization results.

This invention provides a multi-objective relay selection method for energy harvesting cognitive relay networks. In the context of energy harvesting cognitive relay networks, a new multi-objective function is constructed to solve discrete optimization problems by comprehensively considering network energy efficiency and signal-to-noise ratio. The multi-objective relay selection method is then rapidly obtained through a multi-objective quantum monarch butterfly optimization mechanism. This method maximizes network energy efficiency while ensuring the signal-to-noise ratio, providing a new idea and method for solving relay selection schemes in energy harvesting cognitive relay networks.

This invention relates to the field of valve detection technology, and more particularly to a method for detecting internal leakage in valves based on acoustic emission signals. The method includes: acquiring the topology and geometric parameters of the pipeline network to determine the pipeline location characteristics of each valve; based on these characteristics, analyzing the propagation characteristics of acoustic emission signals in the pipe segment where the valve is located for at least two preset frequency bands, and defining the effective monitoring arc segment for each valve in different frequency bands; establishing a dual-modal coverage matrix between the valve and candidate sensor locations, combined with the background noise level; constructing a discrete optimization function based on this matrix to obtain the sensor installation location combination; deploying sensors to collect signals, extracting multi-frequency band time-frequency domain features, and constructing a dual-modal feature vector; combining the feature vector and the coverage matrix to calculate the probability that each valve is a leakage source, and identifying the valve with internal leakage. This invention achieves accurate detection and location of internal leakage in valves by optimizing sensor layout using a dual-modal coverage matrix and fusing multi-frequency band features.

The invention discloses a discrete combination optimization control method of a switchable series-parallel current limiting circuit based on an IGBT and a contactor, and relates to the technical field of power electronic control and electrical testing, the control method comprises the following steps: S1, a real-time sensing step: collecting a target current limiting value and a bus voltage of a system, and real-time state data of each reconfigurable power module, wherein the real-time state data at least comprises a current value, a voltage value and a device temperature. According to the discrete combination optimization control method of the switchable series-parallel current limiting circuit based on the IGBT and the contactor, by dynamically sensing a real-time working condition and constructing a discrete optimization model containing multiple targets, a scheme meeting static and dynamic performance requirements can be automatically and accurately screened out from massive possible topologies; the problems that a traditional fixed topology circuit is limited in adjusting range and insufficient in precision are solved, and the response reliability and the control accuracy of the system under the wide-range working condition are ensured.

The invention belongs to the technical field of data processing, and provides a power market price prediction method and system based on two-stage optimization, and the method comprises the steps: determining an initial generating capacity node value and an initial unit price node value of each generator set, and carrying out the interpolation refining operation of the initial generating capacity node value and the initial unit price node value, obtaining a refined generating capacity node value and a refined unit price node value; establishing an output interval prediction 0-1 planning model of the generator set; performing discrete optimization solution on the output interval prediction 0-1 planning model to obtain a predicted generating capacity interval of each generator set; according to the predicted generating capacity interval, the refined generating capacity node value and the refined unit price node value, establishing an average electricity price prediction quadratic programming model; and performing continuous optimization solution on the average electricity price prediction quadratic programming model to obtain an average electricity price prediction value. The scheme can adapt to market dynamic changes, and prediction precision, flexibility and reliability are effectively improved.

Techniques are described for processing measurement information to select a respective label for multiple voxels or pixels of a medical image dataset. Reference information is described comprising a respective reference dataset for each one of the possible labels, and each label is associated with a material parameter. A respective label for the respective voxel or pixel is selected by using a discrete optimization algorithm to optimize a cost function comprising a first partial cost function and a second partial cost function, and the incorporation of the second partial cost function into the cost function introduces a coupling between the labels at different voxels or pixels.

Techniques for configuring an entity are disclosed. The techniques include generating, based on multiple sets of features inputted into a machine learning model, predictions of multiple sets of performance metrics for multiple configurations of the entity. The techniques also include calculating a change in efficiency between each configuration in the multiple configurations and the base configuration based on a first set of performance metrics for the base configuration and a second set of performance metrics generated by the machine learning model for the configuration. The techniques further include determining, based on values of the change in efficiency calculated for the multiple configurations, a second set of configuration values with a highest increase in efficiency over the base configuration. The techniques additionally include causing the second set of configuration values to be outputted in association with the entity.

The invention discloses a fine-grained visual classification method and system based on adaptive attention quantization Transform. The system comprises an adaptive quantization selection module, a background elimination module and a dynamic hybrid optimization module. The self-adaptive quantization selection module dynamically selects attention scores of a plurality of attention heads by introducing a global and learnable threshold value and adopting a binary quantization mode, selects the most distinguishing feature, and effectively reduces the dilution of attention of the attention head focused on a non-distinguishing mark on a distinguishing area. And meanwhile, discrete optimization is carried out by adopting a straight-through estimator, so that gradient back propagation is realized. And the background elimination module removes features containing background information, so that interference of background noise is eliminated. And the dynamic hybrid optimization module dynamically optimizes and integrates selection and elimination branches so as to realize an optimal classification effect. Experiments prove that the method achieves the current advanced performance on the fine-grained visual classification task.

The invention discloses a traffic signal control optimization method, system, equipment and medium, and relates to the technical field of traffic management, and the method comprises the steps: obtaining vehicle trajectory data, and carrying out the flow reduction processing of the trajectory data, and obtaining the traffic flow; on the basis of traffic flow, signal timing and vehicle speed parameters are introduced, and delay of signals at different intersections is generated in combination with a support vector regression algorithm; the total delay time is minimized by adjusting the phase difference of signals at different intersections, the particle swarm optimization algorithm is improved by introducing dynamic target weight adjustment, the weight ratio between delay and traffic efficiency is adjusted according to the evolutionary process of particles and the current search condition, and the optimal traffic signal control scheme is obtained. According to the algorithm, through a dynamic weight adjustment mechanism and a particle forgetting mechanism, the convergence speed and the global search capability of a phase difference optimization process can be improved, and the defects of a traditional PSO algorithm in a discrete optimization problem are overcome.

The invention discloses a contact-bending performance collaborative optimization design method of a face gear pair under consideration of spatial constraint, which comprises the following steps: step 1, defining a design variable feasible region: determining a design variable optimization space of the face gear pair based on spatial constraint conditions, design variables including continuous optimization variables and discrete optimization variables; 2, a multi-objective optimization objective function is constructed, wherein the minimum tooth surface maximum contact stress and the minimum tooth root maximum bending stress serve as optimization objectives; step 3, configuring a multi-objective optimization algorithm, realizing global performance optimization of the face gear pair through collaborative optimization of spatial constraint and bearing performance, and solving to obtain optimal design parameters; and 4, designing the face gear pair based on the optimal design parameters. According to the contact-bending performance collaborative optimization design method of the face gear pair under the consideration of the spatial constraint, the global performance optimization of the face gear pair design is realized by establishing a collaborative optimization mechanism of the spatial geometric constraint and the multi-target bearing performance.

The present invention discloses an electric CPS intrusion detection system and method based on multi-objective variable-length CNN discrete optimization. The pre-processed historical monitoring data of the electric information-physical system is used as the input data set, and the number of convolution modules, model topology, architectural parameters of each convolution module and batch size, learning rate, optimizer type, weight regularization and other hyperparameters used to construct the convolutional neural network model are variable-length discretely encoded. The indicators of the CNN model on the validation set and the number of model floating-point operations are used as optimization targets. A multi-objective discrete optimization method is designed to perform multi-objective parallel optimization on the CNN model architecture and parameters based on the variable-length convolution module, thereby obtaining a Pareto optimal CNN model that takes into account both model performance and model complexity. For the real-time monitoring data of the electric CPS, the optimal CNN model is deployed online, thereby realizing online intrusion detection of the electric CPS. The present invention reduces the complexity of the model while ensuring high-precision performance indicators of the model.

The application relates to a causal DAG discovery method for an online service system based on fusion of soft priori. The method comprises the following steps: obtaining observation data and text meta-knowledge of the service system, preprocessing to form a standardized sample set, identifying variable types and semantics and outputting; generating a natural language description according to the variable semantics, querying a large language model for an ordered variable pair, analyzing to obtain three types of causal probability vectors, and calibrating to obtain an edge-level priori probability. An appropriate conditional independence test method is selected, high-confidence independent / dependent sentences are divided, and weights are assigned. A candidate directed acyclic graph is selected as an initial structure, parameters are estimated by linear regression, and data fitting scores are calculated, language priori scores, conditional independence penalty terms and counterfactual self-consistency penalty terms are calculated. Fusion is carried out into a hybrid score function, discrete optimization is carried out under the constraint of a directed acyclic graph, and a causal graph structure with the optimal score is output. The method can improve the efficiency and accuracy of a smart operation and maintenance system.

There is provided a non-transitory computer-readable medium storing a calculation program for causing a computer to execute a process. The process includes, an embedding vector generation process for generating an embedding vector that represents each similarity of each problem for multiple problems, in optimization using a continuous relaxation simulated annealing method, which searches for a solution by incorporating continuous relaxation to discrete optimization problems, and a search process for introducing a loss term based on the embedding vector into a loss function in which each element of a matrix obtained by relaxing discrete variables to be optimized into a continuous matrix becomes a discrete optimization problem, and searching for solutions of the multiple problems in parallel.

This application relates to a model construction method, a multivariate time series anomaly detection method, and related components, belonging to the field of system anomaly detection technology. The method includes: initializing learnable embedding vectors for device nodes; determining hyperedges of a first hypergraph structure based on the similarity between learnable embedding vectors; adjusting the device nodes and hyperedges in the first hypergraph structure based on an approximate discrete optimization mechanism to obtain a second hypergraph structure; modeling the learnable embedding vectors, the second hypergraph structure, and the multivariate time series to generate a spatiotemporal hypergraph neural network model; combining the spatiotemporal hypergraph neural network model and a fully connected neural network to generate an initial anomaly detection model; and training and validating the initial anomaly detection model using a target dataset and a mean squared error loss function to generate a target anomaly detection model. This application can improve detection efficiency and accuracy.

The present invention proposes an automatic high-pressure die-casting gate design method based on computer simulation, comprising: obtaining a two-dimensional projection of a three-dimensional model, including: projecting the three-dimensional grid corresponding to the casting onto six planes in a Cartesian coordinate system with the positive and negative directions of the x, y, and z axes as normal vectors; gate parametric design, including: automatically defining the location for placing the gate on the three-dimensional grid corresponding to the casting; gate design optimization, including: generating a parametric gate using a discrete optimization algorithm based on die-casting simulation, including: first randomly generating a series of gate designs, and then using the above-generated gate designs for die-casting simulation calculations; generating a new design using an optimization algorithm based on the final air entrainment corresponding to different designs calculated by simulation to reduce the air entrainment; and after performing a series of iterations, selecting the design with the smallest air entrainment as the final design.

The present application relates to the technical field of unmanned cluster cooperative task allocation, and particularly relates to an unmanned cluster task allocation method based on an improved swarm intelligence algorithm, comprising the following steps: S1, modeling the unmanned cluster cooperative task allocation as a multi-objective discrete optimization, and constructing a corresponding multi-objective optimization model; S2, using an improved multi-objective pelican optimization algorithm to solve the multi-objective optimization model, and the improvement of the multi-objective pelican optimization algorithm comprises: introducing an external archive strategy, a fast non-dominated sorting mechanism and a crowdedness comparison operator; the unmanned cluster task allocation method based on the improved swarm intelligence algorithm, by using the improved multi-objective pelican optimization algorithm as a solver and introducing the external archive strategy, the fast non-dominated sorting mechanism and the crowdedness comparison operator, the present application effectively solves the defects of slow convergence speed, easy to fall into a local optimal solution, insufficient diversity of solutions and weak adaptability of traditional swarm intelligence algorithms when solving multi-objective discrete problems.

The present invention relates to a method, device and storage medium for solving cold chain transportation optimization with processing time, which is applied to the field of cold chain logistics technology. The method comprises: establishing a route optimization model and limiting the constraints of the model, solving the problem with the minimum sum of production and transportation time and the maximum sum of product profits as constraints, obtaining the initial solution of the population through a random search algorithm, and setting irrelevant solutions in the initial solution as non-dominated solutions; optimizing the solution through the optimization coefficient. The solution adopts some ideas of the differential evolution algorithm, combines the characteristics of discrete optimization and multi-objective optimization, and proposes an optimization coefficient with the characteristics of good convergence and simple implementation. When various constraints are met, the transportation and production plans are automatically calculated and generated, thereby achieving a balance between production time and profit, and providing managers with comprehensive decision support, which is no longer limited to transportation efficiency.

There is provided a non-transitory computer-readable medium storing a calculation program for causing a computer to execute a process. The process includes, in a cost function in a search process that performs a search by incorporating continuous relaxation into a discrete optimization problem, in which each element of a matrix obtained by relaxing discrete variables to be optimized into a continuous matrix is a solution of a plurality of discrete optimization problems, outputting a solution of a discrete optimization problem using solutions of the plurality of discrete optimization problems obtained by training a machine learning model by applying a perturbation to the plurality of discrete optimization problems.

The invention discloses a power grid confrontation and defense method and system based on robust-lightweight Transform model optimization, and the method comprises the steps: taking hyper-parameters, such as the number of attention, the type of an activation function, the number of layers, the learning rate and the discard rate, in Transform-Ender as decision variables; a model with good multi-classification detection benign sample F1 score performance in the optimization process is used as an adversarial attack replacement model, an adversarial sample is generated based on a projection gradient descent attack, the number of model parameters after adversarial training is introduced, and minimization of multi-classification detection benign sample 1-F1 scores and an adversarial attack success rate is used as optimization targets. And designing a multi-target discrete optimization technology based on an elite individual external archiving mechanism to carry out offline optimization iteration, and carrying out online deployment on the obtained robust-lightweight Transform model so as to realize confrontation and defense of the intelligent power grid intrusion detection system.

The invention provides a multi-unmanned ship aggregation trajectory planning method based on distributed optimization and a multi-modal network, and belongs to the technical field of unmanned cluster system collaborative decision. The method comprises the steps that an unmanned ship cluster receives an aggregation instruction through an Ethernet communication protocol, and a preset geographic network communication protocol coverage area serves as an aggregation target area; constructing an undirected graph of the unmanned ship cluster and a corresponding Laplacian communication matrix; taking the distance between the current position and the initial position of each unmanned ship as a cost function; based on the undirected graph of the unmanned ship cluster and the corresponding Laplacian communication matrix, coupling the position constraint of each unmanned ship, and solving the optimal aggregation position of each unmanned ship through a distributed discrete optimization algorithm; and based on the kinematics state model of the unmanned surface vehicle in combination with a quadratic Bezier curve, planning an aggregation trajectory of each unmanned surface vehicle from the current position to the optimal aggregation position. According to the method, the communication efficiency is ensured, and meanwhile, efficient collaborative aggregation of unmanned ship clusters is realized.

The present invention provides a method and system for optimizing earthwork allocation based on a discrete marine predator algorithm. The method comprises: initialization, in which the earthwork allocation plan is encoded in triple form; a high-speed ratio phase, a population clustering strategy is applied to the initialized allocation plan; a constant-speed ratio phase, in which half of the prey perform a small-scale movement using Lévy flight, while the other half update their positions using the predator's large-scale movement; a low-speed ratio phase, in which the prey utilize the predator's position to perform a small-scale movement; and considering the fish aggregation effect caused by environmental changes, probabilistic large-neighborhood jumps are performed during the iteration process to avoid falling into local optimal solutions. The present invention's discrete marine predator algorithm, based on a population exchange and update mechanism using triple encoding, outperforms traditional ant colony algorithms in both solution performance and accuracy, providing a novel design approach for converting continuous optimization algorithms into discrete optimization algorithms for solving earthwork allocation optimization problems.

This application relates to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for optimizing parameters of a hydrogen fuel nozzle. The method includes: acquiring a set of structural parameters; classifying each parameter combination using a first classification model and various second classification models, determining the first discrete index classification result and the second discrete index classification result corresponding to each parameter combination; determining the parameter analysis result for each parameter combination using a preset parameter analysis algorithm and the first and second discrete index classification results; determining the target parameter and its corresponding target parameter range among various types of structural parameters based on a preset discrete optimization objective and the contribution degree and direction in the parameter analysis results; and optimizing the target parameter combination based on the target parameter range and a preset continuous evaluation index to obtain the optimized target parameter value. This method can improve the accuracy of parameter optimization.

The invention discloses a region coverage discrete optimization and attitude maneuver planning method for a satellite with a rotary table, and belongs to the technical field of satellite application and observation planning. According to the method, through regional grid division and candidate strip set generation in any direction, an optimized strip capable of achieving the maximum regional coverage rate with the minimum number is obtained through integer programming modeling solution. According to the method, a two-dimensional turntable rotation angle sequence meeting pointing constraint and speed parallel constraint is planned, and a body attitude quaternion trajectory of satellite execution stripe push-broom imaging is obtained through derivation based on minimum Euler rotation. And for attitude pointing switching between strips, quaternion spherical linear interpolation is carried out on an initial quaternion and a target quaternion by designing an Euler rotation angular velocity trajectory, and a continuous rapid attitude maneuver trajectory is obtained. According to the region coverage discrete optimization and attitude maneuver planning method, generation of optimization coverage strips in any direction is allowed, region coverage attitude maneuver gaps are reduced, and observation task execution efficiency is improved.

This invention discloses a transient coordinated control method for solid waste thermal storage and power grid frequency regulation commands. This method, for the first time, proposes a transient coupling model of heat flux density field-phase change excitation, overcoming the constraint of thermal inertia. Through cross-domain integration of discrete optimization algorithms, field strength allocation technology, and holographic simulation technology, it solves the time-domain mismatch problem and achieves transient coordinated control of solid waste thermal storage and power grid frequency regulation commands. The proposed transient coordinated control method for solid waste thermal storage and power grid frequency regulation commands aims to establish a transient coupling model between power grid commands, heat flux density field, and phase change excitation. Through micro-domain phase change excitation and virtual pre-simulation compensation mechanisms, it reduces the thermal response delay from the 300-second level to the sub-second level, improves temperature control accuracy by 50 times, breaks through the theoretical limit of 95% energy conversion efficiency, and achieves a power grid frequency regulation command matching degree of 99.3%. This effectively solves the problems of slow thermal response speed, low energy conversion efficiency, and poor matching degree with power grid commands in existing technologies.

This invention discloses a renewable energy data clustering method and system based on optical quantum computing. First, a renewable energy data clustering optimization problem based on a quadratic unconstrained binary optimization (QUBO) model is constructed based on a discrete optimization objective function. Then, the objective function is matrix-transformed, and the kernel method is used to optimize the clustering of the objective function terms, converting them into matrix form to construct a new matrix-transformed clustering model. Finally, the new QUBO optimization problem is solved using CIM, and the result is transformed into the optimal clustering result. This invention's method, by constructing a QUBO model and combining it with a kernel method to improve data similarity measurement, transforms the renewable energy data clustering problem into a form suitable for CIM solving, thus avoiding the computational bottleneck of classical computational methods in high-dimensional data processing. This method not only preserves the nonlinear characteristics of the data but also achieves efficient clustering optimization computation, significantly improving computational efficiency and clustering accuracy.